《数据挖掘实验四贝叶斯决策分类算法》由会员分享，可在线阅读，更多相关《数据挖掘实验四贝叶斯决策分类算法（5页珍藏版）》请在金锄头文库上搜索。
1、实验四、贝叶斯决策分类算法学院计算机科学与软件学院实验目的：（1）熟悉VC+编程工具和朴素贝叶斯决策算法。（2）对AllElectronics顾客数据库查询得到先验概率和类条件概率。（3）在样本集上用VC+编程工具编写用朴素贝叶斯算法分类的程序，对任 务相关数据运行朴素贝叶斯分类算法，调试实验。（4）写出实验报告。实验原理：1、先验概率和类条件概率先验概率：先验概率定义为训练样本集中属于Ci类的样本（元组）数N.与尸（G）冷总样本数N之比，记为，。类条件概率：类条件概率定义为训练样本集中属于q类中的具有特征X的 样本（元组）的个数ni与属于q类的样本（元组）数Ni之比，记为 p（xc）*;。2
2、、贝叶斯决策贝叶斯决策（分类）法将样本（元组）分到Ci类，当且仅当I m i * J 对 1jm，jHi其中，训练样本集中的样本（元组）可被分为m类。 该算法流程图如下： 实验内容1、实验内容用贝叶斯分类器对已知的特征向量X分类：1）由AllElectronics顾客数据库类标记的训练样本集（元组）编程计算先验 概率P（Cj）和类条件概率P（X|Cj）,并在实验报告中指出关键代码的功能和实现方 法；2）应用贝叶斯分类法编程对特征向量X分类，并在实验报告中指出关键程 序片段的功能和实现方法;3) 用检验样本估计分类错误率2、实验流程图3、关键代码1、定义存储结构class Date public
3、:string age;string income;string student;string credit;string buy;void print() credit cout age income student buyname1;infile.open(name1,ios:in);if(infile.fail()cout error open! endl;3、计算类条件概率(通过计算累加和来计算) coutage:iage;coutincome:iincome;coutstudent:istudent;coutcredit:icredit;for(int k = 0;kdatesize
4、;k+) if(datek.age=iage&datek.buy=yes) agey+; if(datek.age=iage&datek.buy=no) agen+; if(datek.income=iincome&datek.buy=yes) incomey+; if(datek.income=iincome&datek.buy=no) incomen+; if(datek.student=istudent&datek.buy=yes) studenty+; if(datek.student=istudent&datek.buy=no) studentn+; if(datek.credit=
5、icredit&datek.buy=yes) credity+; if(datek.credit=icredit&datek.buy=no) creditn+; p3=(float)agey/(float)y; p4=(float)agen/(float)n; p5=(float)incomey/(float)y; p6=(float)incomen/(float)n; p7=(float)studenty/(float)y; p8=(float)studentn/(float)n; p9=(float)credity/(float)y; p10=(float)creditn/(float)n
6、; px1=p3*p5*p7*p9; px2=p4*p6*p8*p10;px3=px1*p1;px4=px2*p2;coutP(age = iage|buy = yes =agey/y=p3endl;coutP(age = iage|buy = no = agen/n=p4endl;coutP(income = iincome|buy = yes = incomey/y=p5endl;coutP(income = iincome|buy = no = incomen/n=p6endl;coutP(student = istudent|buy = yes =studenty/y=p7endl;c
7、outP(student = istudent|buy = no = studentn/n=p8endl;coutP(credit = icredit|buy = yes =credity/y=p9endl;coutP(ctedit = icredit|buy = no = creditn/n=p10endl;coutP(X|buy = yes) = px1endl; coutP(X|buy = no) = px2endl;coutP(X|buy = yes)P(buy = yes) = px3endl;coutP(X|buy = no)P(buy = no) = px4px4)coutvv朴
8、素贝叶斯预测buy = yesendl;elsecoutvv朴素贝叶斯预测buy =noendl;system(PAUSE);return 0;1. 实验数据RIDageincomestudentcredit rati ngChss: buys_compute-1youthhighnofairno2youthhighnoexcellentno3middle agedhighnofairyes4seniormediumnofairyei5seniorlowyesfairyes6seniorlowyesexcellentno7middleagedlowyesexcellentyesSyouthmediutnnofairno9youthlowyes-fairyes10seniormediumyesfairves J11youthmediumyesexcellentyes12Twiddle agedmediumnoexcellentyes13 Twiddle agedliihyesfairyesseniormediumnoexcellentno实验结果：用训练样本集中元组进行测试:用未知数据测试:C: XFrograM F ilea Micro soft Visual S-tudioMyPrD jcct 3da4DcbiiEla4. czg
